AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Boosted Attention: Leveraging Human Attention for Image Captioning

Shi Chen, Qi Zhao

ECCV 2018 · Vision-language and multimodal › Human and embodied modeling · approach: Concept-guided image captioning · 53 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add stimulus-based attention (Boosted Attention) (ablation)Soft-attention ResNet-101 captioner · MSCOCO Karpathy testCIDEr: 1.062 -> 1.122Improved
add stimulus-based attention (Boosted Attention) (ablation)Soft-attention ResNet-101 captioner · MSCOCO Karpathy testBLEU-4: 0.335 -> 0.354Improved
add stimulus-based attention (Boosted Attention) (ablation)Soft-attention ResNet-101 captioner · Flickr30K testCIDEr: 0.523 -> 0.586Improved
add stimulus-based attention (Boosted Attention) (ablation)Soft-attention ResNet-101 captioner · Flickr30K testBLEU-4: 0.267 -> 0.274Improved
add stimulus-based attention (Boosted Attention) (ablation)Soft-attention ResNet-101 captioner · Flickr30K testMETEOR: 0.197 -> 0.208Improved
add pretrained saliency weights to stimulus-based attention module (BAM* to BAM) (ablation)Boosted Attention · MSCOCO Karpathy testCIDEr: 1.111 -> 1.122Improved
add pretrained saliency weights to stimulus-based attention module (BAM* to BAM) (ablation)Boosted Attention · Flickr30K testCIDEr: 0.571 -> 0.586Improved
add pretrained saliency weights to stimulus-based attention module (BAM* to BAM) (ablation)Boosted Attention · MSCOCO Karpathy testMETEOR: 0.262 -> 0.265Improved
add architecture with randomly initialized stimulus module (BAM*) over baseline (ablation)Soft-attention ResNet-101 captioner · MSCOCO Karpathy testCIDEr: 1.062 -> 1.111Improved
ensemble of 4 BAM models vs ensemble Att2all on online serverBoosted Attention · MSCOCO online test (C5)CIDEr: 1.147 -> 1.083Hurt
use stimulus-based attention alone as captioning attentionStimulus attention · Karpathy test attention mapscorrelation with captioning attention: CC = 0.222No clear effect
compare top-down attention maps with stimulus-based mapsBoosted Attention · Karpathy test 5000 imagesCC and Spearman: CC = -0.256, Spearman = -0.369Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy